waders and their estuarine food supplies - Vlaams Instituut voor de ...

and annually. Second, the lower size acceptance
threshold as well as the depth selection varies being a
function of the intake rate.
Knot feed by touch when they search for benthic
bivalves, but in contrast to Oystercatchers. they ingest
then pre) whole. This makes ihc lun veslablc lids lion
of prey for Knot much smaller than for Oystercatchers
(Fig. 14B). Macoma is a preferred prey for Knot, but
they ignore the size classes smaller than 10 mm long
and reject Macoma larger than 16 mm long: the small
prey are unprofitable, while the large ones are too wide
to be swallowed (Zwarts & Blomert 1992).
Knot visited our Stud) site only in August when the
total biomass of Macoma over ten years varied between
6 and 35 g nr 2 , with an average of 17.7 g nv 2
(Fig. 15). On average. 44',{ of this biomass belonged to
the suitable size classes, whereas only 31% was
harvestable. i.e. both of suitable size and accessible
(living in the upper 2 cm). The annual variation of 6 to
35 g m - in the total biomass of Macoma was small
compared to those of other benthic species (Beukema
et al. 1993). the standard deviation of 8.6 g nr- being
only 49% of the mean (17.7 g in 2 ). The relative
standard deviation (RSD) lot suitable biomass (6 to 16
mm long) was 44' i. and thus lower than that for the
total biomass. The annual variation in the fraction of
Macoma living within reach of the bill was much
larger still: for instance. 98% of the prey was found in
the upper 2 cm of the substrate in August 1984. against
only 15% in August 1986 (Fig. 15). On average. 54%
was accessible and the RSD was 60%. As a consequence
of this large variation in prey accessibility,
the RSD of the harvestable biomass increased still
further to 77%. It was therefore the variation in depth
distribution that was a major contribution to year-toyear
fluctuations in the biomass of Macoma actually
harvestable by Knot staging in our study area in
August. Piersma et al. (1993b) arrive at the same
conclusion in their study of Knot staging on Griend.
western Dutch Wadden Sea. Knot did not stage in our
study area when the biomass of the harvestable
Macoma was low (Fig. 15).
The harvestable prev fraction in touch-feeding
waders, such as the Knot, is less complex to measure
than in most waders that feed by sight. Figure 14C
illustrates the relatively simple situation of a Curlew
searching for the siphon holes of Mya. This prev is
FOOD SUPPLY HARVESTABLE BY WADERS
n
1977 78 79 80 81 82 83 84 85 86
Fig. 15. The biomass of Macoma balihica in August 1977 to 1986,
given for all size classes i 'total' i. lor onlv the specimens III ihe range
6 lo 16 mm I 'suitable si/e' i and lor animals of 6 lo 16 mm living in
Ihe upper 2 cm of the substrate I 'harvestable' I. The grey field gives
the averages* SD. The lower panel slums the response of the Knot
(peak numbers in the study area). Dala from Zwarts el ul 11992).
harvestable if it lives within reach of the bill (13 to 16
cm), if il is profitable (size > 3 cm) and. at least for
Curlews feeding by sight, if the siphon hole is v isible
at the surface (Zwarts & Wanink 1984). Only a small
part of the profitable fraction is actually accessible.
This is probably the main reason whv short-billed
male Curlews (bill length 10 to 13 cm) never feed on
Mya, while it is the main prey for the females (bill
length 13 to 16 cm) in areas where ihe prev specieoccurs
(Zwarts & Wanink 1984). The detectable
fraction also varies considerable. Siphon holes arise
when Mya extend their siphon lo the surface for
suspension feeding. Mya cannot feed at low tide as no
water lies on the surface, but the siphon holes may
remain visible until they decay over the low water